[03:47:38] 10Machine-Learning-Team, 10Gerrit, 10Release-Engineering-Team: gerrit: scoring/ores/editquality takes a long time to git gc - https://phabricator.wikimedia.org/T237807 (10thcipriani) [03:48:42] 10Machine-Learning-Team, 10artificial-intelligence, 10Diffusion, 10Release-Engineering-Team, 10editquality-modeling: Gerrit repo scoring/ores/editquality not mirroing - https://phabricator.wikimedia.org/T224996 (10thcipriani) [03:55:12] 10Machine-Learning-Team, 10Diffusion, 10Release-Engineering-Team, 10Wikimedia-GitHub: LFS objects are not mirroring from Github through Phab to Gerrit consistently - https://phabricator.wikimedia.org/T212818 (10thcipriani) [03:55:31] 10Machine-Learning-Team, 10ORES, 10Release Pipeline, 10Release-Engineering-Team, 10SRE: Execution of the deployment pipeline should be configurable via .pipeline/config.yaml - https://phabricator.wikimedia.org/T210267 (10thcipriani) [03:58:31] 10Machine-Learning-Team, 10Release-Engineering-Team, 10Performance Issue: Investigate PEP 503 repo for production deployment of python wheels - https://phabricator.wikimedia.org/T192478 (10thcipriani) [07:28:37] TIL https://packages.debian.org/source/bullseye/kubernetes [07:39:32] the current plan for Bullseye is that kubernetes in Debian stable will only ship the client and the full k8s suite will be provided via fasttrack.debian.net (since k8s requirements for the Go compiler change too frequently to catch up in a stable release) [07:40:30] but that's no real downside, the packages on fasttrack are managed by Debian developers like the packages in stable with the same standards [07:41:11] k8s in fasttrack will follow each kubernetes release series until it's EOLed and replaced by the next one [07:44:56] that seems a great news! [07:45:33] yeah, I think it's a good compromise! [07:45:34] I didn't check with Alex but do we have the plan to use fasttrack.debian.net for k8s 1.20? [07:46:10] we (as ML) will need k8s 1.19+ sooner or later, all the fancy kubeflow stuff runs only on latest and greatest of course [07:46:30] this new approach is really fresh, it was only discussed in Debian two days ago and there's not even a public reference at this point [07:47:56] can best be sorted on a Phab task with all stakeholder I guess? [07:48:13] ahhh okok [07:48:33] yes of course, I thought it was a more consolidated thing [07:48:38] if there's anything in the packages missing (build flags/options/whatever), also happy to relay feedback to Janos and get it changed [07:48:58] I'll bring this up in #serviceops as well [07:49:29] or a task is better, I'll report this conversation [07:49:30] :) [07:49:30] there was a previous attempt to ship k8s in Debian, but it followed a strict "package every dep in the distro" attempt and was doomed to fail since k8s is too big and moves to fast [07:54:32] 10Machine-Learning-Team, 10serviceops: Kubernetes packages in Debian Bullseye - https://phabricator.wikimedia.org/T280625 (10elukey) [07:54:34] opened a task :) [08:05:10] 10Machine-Learning-Team, 10SRE, 10serviceops: Kubernetes packages in Debian Bullseye - https://phabricator.wikimedia.org/T280625 (10MoritzMuehlenhoff) [08:49:30] 10Machine-Learning-Team, 10SRE, 10serviceops: Kubernetes packages in Debian Bullseye - https://phabricator.wikimedia.org/T280625 (10akosiaris) For the "main" set of clusters, we have devised a plan to adopt upstream binaries but without relying on their repos. The Policy as well as the reasoning for that (th... [09:42:25] 10Lift-Wing, 10Machine-Learning-Team: Install Istio on ml-serve cluster - https://phabricator.wikimedia.org/T278192 (10elukey) I was able to bootstrap minikube with k8s 1.20.2 (the other ones failed for cgroup issues..) ` elukey@wintermute:~/Wikimedia/scratch-dir/istio-1.6.2$ ./bin/istioctl operator init Usin... [10:16:57] 10Lift-Wing, 10Machine-Learning-Team: Install Istio on ml-serve cluster - https://phabricator.wikimedia.org/T278192 (10elukey) Even if used istioctl for this use case (and not helm), we should have a complete list of Docker images to add to our internal registry. In theory the best thing would be to avoid pull... [12:11:56] 10Lift-Wing, 10Machine-Learning-Team: Install Knative on ml-serve cluster - https://phabricator.wikimedia.org/T278194 (10elukey) Tried to install knative + istio following https://github.com/kubeflow/kfserving/blob/master/test/scripts/run-e2e-tests.sh#L75-L102 on minikube + k8s 1.20.2 (1.16.0 seems not running... [12:21:10] 10Lift-Wing, 10Machine-Learning-Team (Active Tasks): Install certmanager on ml-serve cluster (if needed) - https://phabricator.wikimedia.org/T280661 (10elukey) [12:37:09] 10Lift-Wing, 10Machine-Learning-Team: Install Istio on ml-serve cluster - https://phabricator.wikimedia.org/T278192 (10elukey) Addendum - the istio operator pod is needed only if we want to support istioctl, it seems not needed when using helm. As starting point, we could try to import `istio/proxyv2` and `ist... [16:45:32] 10Machine-Learning-Team, 10Discovery-Search, 10Epic, 10Growth-Team (Current Sprint): [EPIC] Growth: Newcomer tasks 1.1.1 (ORES topics) - https://phabricator.wikimedia.org/T240517 (10Rileych)